The overall theme of this new project is to establish and validate blood- and imaging-based biomarkers associated with the risk and progression of late onset Alzheimer's disease (LOAD), mild cognitive impairment (MCI) and the rate of cognitive decline in late life. We also propose to develop a framework with which we can understand how biomarkers interact and how they fit into the temporal sequence from healthy aging to dementia. This proposal is built on two decades of epidemiological research and systematic data collection in the multi-ethnic, Washington Heights, Hamilton Heights, Inwood, Columbia Aging Project (PO1AG07232). Over the past 20 years, we have investigated the rates of LOAD, MCI and cognitive decline in this urban community in northern Manhattan. We have investigated environmental, health-related and genetic risk factors of disease and predictors of disease progression by collecting longitudinal data on cognitive performance, emotional health, independence in daily activities, blood pressure, anthropometric measures, cardiovascular status and selected biomarkers in this elderly, multi-ethnic cohort, including lipids, amyloid peptides, sex hormones, homocysteine, insulin and C-reactive protein (CRP), and MRI. We have reported that the rates of disease and the frequency of disease risk factors vary across ethnic groups. We have identified one of the largest, multi-ethnic groups of incident LOAD cases facilitating studies of disease progression. Clinical and genetic data as well as biological resources are present for several thousand individuals. Biomarkers, cellular, biochemical or molecular alterations measurable in human tissues, cells, or fluids or by radiological means, are typically chosen because they are directly or indirectly in the causal pathway of disease. The emergence of structural and functional brain imaging has revolutionized epidemiological studies, particularly those using biomarkers for Alzheimer's disease. Positron emission tomography brain imaging using 11C Pittsburgh compound B is considered an in vivo measure of brain amyloid plaque load, while structural MRI, especially changes in brain volume and cerebral blood flow (CBF), can be considered an in vivo measures of neurodegeneration. In this new proposal, we will focus this longitudinal investigation on two sets of blood biomarkers that not only show consistent and robust associations to the risk of LOAD, MCI and cognitive decline, but that address the putative mechanisms related to amyloid burden and insulin resistance. We will take full advantage of the prospective design in this multi-ethnic cohort and the clinical, biological and brain imaging data collected to address six major hypotheses. The first two primary specific aims consider blood biomarkers as not only predictors of cognitive decline, MCI, LOAD and LOAD progression, but also as intermediate steps in the disease pathway, including neurodegeneration and cerebrovascular burden (MRI) and amyloid plaque load (PIB). In the last primary specific aim, brain imaging variables are predictors and cognitive decline, MCI, LOAD as well as LOAD progression are main outcomes.